Sunday, October 25, 2026

How Much of Our Revenue Did Marketing Actually Make?

Ask a marketing director how much revenue marketing made last year and you will get a number. Ask the sales director and you will often get a smaller one. Ask the finance director and often he will show a complicated set of emotions depending on how the revenue and EBITDA numbers look at the time:

If they are good, he will sometimes look dazzled, like he’s seen a wonderful magic trick. If the numbers are bad, he will more often than not look frustrated, yet still just as mystified.

All three are reasonable people looking at the same business.

Marketing-sourced revenue is the revenue from deals where marketing made the first qualified contact, under a definition both Sales and Marketing have agreed. Most companies cannot state theirs. That is why three directors give three different numbers.

When I was at a telco software company, our entire universe was about 400 big accounts. Once we landed a mega deal. It had come in through an event, then been nurtured by us in the email channel for nine months. Then it closed.

It turned out the salesperson who closed it was old friends with the man who signed. The relationship went back years, offline of course, so no way of tracking or confirming that. But the deal went to Sales. A cynic may say that this was more the work of politics than attribution.

This has happened many times in my career in B2B software. It happens often enough that HubSpot built a method for it: the Sales and Marketing Service Level Agreement, or SLA. 

Both teams sign one page. Marketing commits to a number of qualified leads a month, defined in writing. Sales commits to contact each one within a set time and work it a set number of times. Every month the two numbers are read out together.

HubSpot’s final ingenious addition is a Judicial review: any disputed deal goes to a C level leader (above the political fray) who runs neither Sales nor Marketing, usually the President, COO or CFO, who looks at the evidence and rules.

I have recommended it at several companies since I first read about it over six years ago. It has never been implemented, and I have never seen it run anywhere, even though it comes closest to solving the problem.

The reason is not complicated. An SLA makes both teams measurable by someone else, and nobody who is winning the current argument wants a referee.

Why is the marketing revenue number usually wrong?

Because there are two systems and two definitions, and nobody owns the join.

Marketing lives in a marketing platform. It knows who opened an email, clicked an ad, came to a webinar. Sales lives in the CRM. It knows who was called, who got a proposal, who signed. The two systems are supposed to talk to each other. 

In my experience, across a dozen B2B companies, Sales and marketing alignment is on a spectrum; never perfect, rarely terrible, but that continuum is flexible and rapidly changing. It needs constant maintenance to keep it on the 'healthy' side.

The first time I saw this properly was at a Fintech software company in 2013, where I ran marketing automation on one system while Sales ran the pipeline on another. A customer could be a hot lead in one and a stranger in the other.

When the systems disagree, each side counts its own way. Marketing counts everyone it touched. Sales counts everyone it closed. Both numbers are true, but neither tell the whole story. 

That’s why in B2B you need constant honest communication (addressing the brutal facts in your funnel) to keep all your numbers healthy: from lead to sale conversion rates, to cost per sale, to Return on Ad Spend, and so on.

What does the real number look like?

Different from both of the ones you were given, and findable depending on the systems you have in place – perhaps immediately, perhaps longer (maybe a week, say, for a poorly integrated platform).

The method is not clever. You take every closed deal, you look up every contact on it in the marketing system, and you write down what marketing did to that person before the deal was opened. Then you do the same for every open lead. It is tedious. It is also the only way to get a number that both directors will sign.

What are three levers I can pull right now to improve my campaign peformance?

None of them needs new budget. Each need one decision from the top and a fortnight of someone's time.

How do you find leads lost between marketing automation and the CRM?

Lever 1: find the customers you have already paid for.

This is the cheapest revenue in the building. The marketing money that produced it has already been spent.

Every company I have worked in had leads that were generated, paid for, and then lost between two databases. The fix is a rule, not a project: once a fortnight, every contact in the marketing system is matched to the CRM, and anything that does not match gets a name next to it.

If you do want the bigger fix, it is worth it. The question to ask: when did we last check that every lead we paid for reached a salesperson? If the answer is a date, good. If it is unclear, or only gives you a partial picture, this is your first lever.

How do you get Sales and Marketing to agree on lead definitions?

Lever 2: make Sales and Marketing count the same thing.

The argument about whose lead it was is a definition problem, not a personality problem. But it is also a big picture systems issue. If we are not clear about definitions, then we are also unclear about incentives, and most importantly, accountability. It has a one-page fix, and it is the SLA from the top of this piece.

Write down once what counts as a marketing-sourced deal and what counts as marketing-influenced. Get both directors to sign it. Then stop arguing about individual deals. Where you can, replace the argument with a score. That is the real test of a definition. Not whether it is fair, but whether the sales team changes its behaviour because of it.

The revenue effect is large. Once both teams count the same way, marketing stops generating volume (‘MQLs’ for their own sake in a performative manner) to win the argument and starts generating leads Sales will work. At a fintech scale-up where I made this change, the cost of a sales demo fell eightfold in a year. Same budget, same team, different definition of a good lead.

How should a B2B company reallocate its marketing budget?

Lever 3: spend where the last ten wins came from.

Most marketing budgets are set by last year's budget. The better question is where the last ten customers actually came from, by name, and whether the money is going there.

I would caveat that by saying – sure, you need to look at the longer-term trends, not least because the large data sets show greater statistical significance. (Bayesian statistics, though, is the primary approach well-suited for working with small data sets – as in your last ten deals).

The deal-by-deal check answers this question. What I see usually is that a handful of behaviours predicts  a won deal far better than the rest: repeat visits to the website, a recent email click, a look at the pricing page. The campaigns that produced those behaviours were not the biggest. They were the ones aimed at a short list of accounts rather than a long list of names.

I have seen this at every size of company. At a telecoms billing software company, a year of work on one named account became a five-million-pound annual contract. At a B2B software company a decade later, moving paid media from broad targeting to a named-account list halved the cost of a lead.

At a martech scale-up, the two largest customers of the year came from targeted outreach to a handful of global brands, not from the campaign with the biggest budget.

The lever is to move money, not add it. Take ten percent from the channel that produces the most leads and put it into the channel that produced the most customers. They are rarely the same channel.

This approach can also work on the campaign level.

The question to ask: where did our last ten (for B2B Enterprise), or a hundred (for smaller transactional selling) customers come from, and does the budget match that list?

Three questions to ask on Monday

You do not need to understand the systems. You need to ask three questions and insist on a date, a name or a number in reply.

1. When did we last check, deal by deal, that marketing's revenue number and sales' revenue number describe the same customers?

2. Do Sales and Marketing have one written definition of a marketing-sourced deal, signed by both, and does Sales use it to decide who to call?

3. Where did our last ten customers come from, by channel (and Campaign), and does the budget match that list?

4. Are you regularly disagreeing over whether a deal 'comes from' sales, or marketing?

If all four come back with a clear answer, you have a commercial engine most companies would envy. If any come back with a pause, you have found your lever. 

And if the answer to the final question is yes, you know why. 

Nobody who is winning the argument wants a referee. 

That is a decision only the person above both teams (President/COO/CFO) can make - Certainly not the Head of Sales, nor the Head of Marketing.

But as I've said many times, most of these problems don't come from tech (but tech is the easiest scapegoat to blame, for sure), they come from politics, systems, rules, accountability (Ownership) and above all, Governance.

Frequently asked questions

What is marketing-sourced revenue? Revenue from deals where marketing made the first qualified contact, under a definition Sales and Marketing have both agreed. Marketing-influenced revenue is wider: any deal marketing touched along the way.

What is a Sales and Marketing SLA? A one-page agreement. Marketing commits to a number of qualified leads a month; Sales commits to how fast and how often each one is worked. Both numbers are reviewed together monthly.

How do you reconcile marketing automation with the CRM? Match every contact across the two systems on a shared identifier, flag the ones that do not match, and give each unmatched record an owner. Fortnightly is enough.

Why do Sales and Marketing disagree about leads? They measure different things. Marketing counts engagement; Sales counts readiness to buy. One scoring model, adopted by Sales (and by ‘adopted’ I mean they need to believe in it, like a priest does in Jesus), closes the gap.

How often do marketing automation and the CRM disagree? More than most leaders expect. Daily, at some companies, weekly at others, rarely just monthly, and never have I seen them never disagree.

What is the fastest way to increase marketing-sourced revenue? Find the leads already paid for and lost between systems, agree one definition of a good lead, and move budget toward the channels that produced the last ten customers.

About the author

Rudy Parker is a Senior Demand Generation Manager with 14 years in B2B SaaS marketing (Cybersecurity/Fintech/Martech/HR/Telecommunications - Software). 

No comments: